Influenza A and B viruses cause a highly contagious respiratory disease in humans. Influenza B viruses infect only humans, whereas influenza A viruses infect many avian and mammalian species and are responsible for periodic pandemics that can result in high mortality rates. H5N1 viruses are candidates for causing a future pandemic: existing H5N1 viruses might require only three mutations to become transmissible between humans, highlighting the need for new influenza virus antivirals. The present research focuses on interactions of several host proteins with particular influenza A and B viral proteins and on how these interactions impact virus replication. Several of these interactions reveal new targets for antiviral development. The NS1 protein of influenza A viruses (NS1A) has important roles in virus infection and virulence. An important goal is to elucidate the poorly understood rol of NS1A in viral RNA synthesis, an endeavor that was facilitated by the purification of macromolecular complexes containing the viral polymerase from infected cells. In addition to the four proteins of viral polymerases (PA, PB1, PB2 and NP), these complexes contain NS1A and specific cellular proteins. Surprisingly, the ZAPL antiviral protein is one of the cellular protein. This association is the end result of a novel host-virus interaction in which viral polymerase proteins counteract a previously unknown antiviral activity of the human ZAPL protein. The new ZAPL activity via its C-terminal domain promotes ubiquitination and degradation of PA and PB2, and this antiviral activity is, in turn, neutralized by PB1 and NP binding to a different region of ZAPL. One aim is to elucidate the mechanism of this novel host-influenza A virus interaction, which involves potential targets for antiviral development. Two cellular RNA helicases, DHX30 and DDX21, are also associated with these complexes. Both helicases bind NS1A and function in temporal regulation of viral gene expression by delaying onset of the high rate of viral RNA synthesis. DDX21 also binds the PB1 protein and inhibits assembly of polymerases, and thus functions as a newly identified host anti- influenza A virus protein that is countered by the NS1A protein. Hence the NS1A-DDX21 interface is a potential target for developing antivirals. The third aim focuses on how the NS1 protein of influenza B virus (NS1B) inhibits the antiviral activity of the interferon-induced ISG15 protein, which can be conjugated to various proteins. NS1B binds only human and non-human primate ISG15s. In cells infected by a newly generated virus whose NS1B protein does not bind human ISG15, the levels of unconjugated NP and M1 in cells and in progeny virus particles are reduced. Because these effects are dependent on ISG15 conjugation, it is likely that they are caused by dominant negative actions of ISG15-conjugated NP and M1. This hypothesis will be tested. To elucidate pathogenic effects of ISG15 and its conjugation, new mice strains in which the mouse ISG15 gene is replaced by the human ISG15 gene are being generated. These results should be relevant to mechanisms by which IFN-induced ISG15 and its conjugation affect infection by other viruses.